Depth extraction is fundamental in computer vision, and is a useful building block in perceptual computing for a variety of applications such as gesture tracking and face recognition. A stereo camera system is one of the techniques used for depth extraction. There remain, however, a number of non-trivial issues and limitations associated with depth extraction techniques.